SI413: Why Trend Following Is More Than Crisis Alpha ft. Andrew Beer & Tom Wrobel

Core argument. Andrew Beer's headline claim: this has been one of the most extraordinary risk-adjusted periods trend following has produced. The SG CTA Index is up 23% over the trailing twelve months at a beta of roughly 0.3 to equities, versus MSCI World's 21% and bonds' 2% — meaning CTAs matched a strong equity market with a fraction of the equity risk. Tom Wrobel corroborates with SG index data through mid-August: SG Trend +0.55% in August (+8.5% YTD), SG CTA Index +0.58% (+8.74% YTD), the short-term index +1.08% (+4.3% YTD), against MSCI World +3.31% (+13.05% YTD), MSCI World ex-US/Canada +2.31% (+12.34% YTD), the S&P 500 TR +3.46% (+13.19% YTD), and the US Aggregate Bond Index +0.40% (+0.06% YTD). Wrobel's trend barometer has fallen to 27% — few markets are currently trending, meaning the opportunity set is narrow even as year-to-date numbers hold up. One live position flagged: SG's model trend portfolio has gone short gold, which Wrobel says is generating "interesting performance" as CTAs adapt to the new regime.

Beer's larger point is that the gains this year came from several uncorrelated themes running at once — long equities (a contrarian, FOMO-driven trade post-"Liberation Day"), gold (long, then exited before it got "crushed"), crude oil (long into the Iran-war squeeze), and a weak yen — rather than one dominant crisis trade. He argues this is the actual mechanism behind trend following's value, not crisis protection: rules-based systems can enter emerging trends early and, critically, exit ruthlessly when a position stops working, something he says human, discretionary allocators structurally struggle to do ("very very hard for humans to give up on their favorite ideas").

Mechanism — why the resilience held. Active, ongoing reallocation across markets, not a static book, is what let CTAs hold onto year-to-date gains through a rough middle of 2026 (a June correction, a soft July) without a single shock doing lasting damage. The clearest test was the yen intervention: Beer expected CTAs to be hit harder given concentrated short-yen dollar exposure, but the space actually outperformed through those days because it was simultaneously long crude oil — a case where one nightmare scenario was offset by an unrelated position.

What has to be true / where the view could break. Beer contrasts CTA-style notional leverage in liquid, margined futures with the leverage used by "Situational Awareness," an AI-themed, venture-style fund that grew from zero to roughly $45bn using an estimated 4:1 leverage, then reportedly fell around 67% in July while still up roughly 80% year-to-date. His point: borrowing against illiquid collateral that can force an unwind is a fundamentally different risk than futures margin overseen actively by the manager — he says he isn't aware of any CTA blow-up of that magnitude in the strategy's history. Both guests separately note a genuine anomaly: in July, two Morningstar/UCITS CTA indices were up while the SG CTA Index was down, possibly due to a couple of large constituents carrying more yen-carry exposure than pure trend, though Wrobel says he'd need to dig further, and flags UCITS commodity settlement lags as a possible confound in day-level comparisons.

Beer is also skeptical of ongoing financial engineering in the space — twenty-five years of added instruments and complexity, he says, "the Sharpe ratio hasn't budged." He separates vehicle innovation (managed accounts, portable alpha, low-cost ETFs, UCITS) — which he supports as legitimate ways to expand the investor base — from investment-side over-engineering, which he doesn't think adds edge. On non-trend/short-term CTA strategies specifically, his own research finds them close to "a zero Sharpe ratio exercise": they look good on a backtest chart (earlier entries, later exits) but overtrading erodes the benefit, and he sees no durable diversification value versus core trend. Wrobel's counter: shorter-term sleeves smooth the return profile enough to survive investment-committee scrutiny during volatile stretches, even if — as with several US mutual funds that pitched faster models after the 2022 whipsaws — that protection carries a real, measurable cost in subsequent performance.

On benchmark construction, both flag a selection-bias problem worth flagging to allocators: 17 of the SG CTA Trend Index's current 20 constituents have outperformed the index over the past ten years, yet only 9 of the original 20 constituents from a decade ago are still in the index today — survivors look strong partly because underperformers get quietly replaced. On a straight-return basis the live index would rank only around the 15th percentile against its current constituents; on a Sharpe basis (which credits diversification across managers) it improves to roughly the 35th percentile. Separately, Beer notes the SG CTA Index's 25-year return equates to roughly cash-plus-250bps net of fees, and Wrobel points out that many index constituents report via managed-account composites that may not include cash returns at all — a rising-rate-era drag on the benchmark that neither has fully quantified.

Framing dispute (not a market call, but a real disagreement). Wrobel notes the industry narrative has shifted from "crisis alpha" toward "risk mitigating strategies" (RMS) language, citing a recent SBAI-style guide. Beer pushes back: he argues "crisis alpha" was itself a marketing framing that emerged because a cash-plus-250bps, 25-year return profile means most years look like near-cash, punctuated by one outsized payoff year — which forces managers to sell the lumpy years as insurance. His preferred fix isn't renaming the strategy but raising its underlying Sharpe ratio so it earns a genuine premium across the cycle rather than needing to be marketed as crisis protection.

A smaller thread: a listener asked about "third-generation" systematic commodity funds. Beer notes his own commodity-fund experience was the opposite model — fundamental, physical-market traders (in the mold of pre-Enron merchant power/gas trading) exploiting information asymmetries against CTA flow, not systematic trend/carry blends. Wrobel classifies the funds the listener meant as "commodity-enhanced beta" — blends of trend, carry, and curve-rolldown strategies built to mitigate the historical cost of staying long commodities (whose 25-year peak-to-trough drawdown Beer puts at roughly 80%, versus a sub-20% max drawdown for the CTA space over the same period).